Investigation into the abundance and distribution of Cepea nemoralis at Pulpit Hill
By: Emma Louise Black, Christopher Tobi Travih, Daniel Guenin, Kamela Bellovoda, Joey Kielty.
Introduction Cepea is a highly polymorphic species of snail. One of the most abundant types of Cepea found at the site is C. nemoralis. The most well-known genetic polymorphism of this species is that of shell pattern. The shell of C. nemoralis has polymorphism for colour and for presence, number and appearance of up to five dark bands. The major classes of colour are yellow, pink and brown. There is much variation in how the bands are expressed which includes, fully pigmented bands, which are common, and interruption or complete loss of the bands. In many of the populations studied, fusions of adjacent bands and spreading of band pigment over the whole of the shell has also been frequently found. Less common are changes in the colour of the band pigment to orange or light yellow. Genetic variation has also been found in shell shape and size, fecundity, viability, developmental time and some aspects of behaviour in some populations (Jones et al 1977).
According to the Jones et al. study on C. nemoralis there are different kinds of selection acting on this species that are very important for the maintenance of polymorphism. There is visual selection which is based on selection by predators. Another form of selection is climate for which the absorption of the energy from the sun varies between dark/light and banded/un-banded shells. There are also frequency dependent selection in which rare alleles are selected over more spread ones. This happens due to the fact that the individuals containing rare morphs are not easily found. Other types of selection important for the maintenance of shell polymorphism are density dependent and disruptive selection (Jones et al 1977).
Genetic drift and selection may alter the distribution and abundance of the different types of C. nemoralis. Our experimental design may distinguish the effects of drift from selection. We have sampled in each of the different environments (woodland, open grass and hedges) at heights differing of 10 metres above and below the transect line. If there is random distribution of the different types of snail and there is no distinct pattern in their distribution, then we can suggest it is due to genetic drift as every species is subjected to drift which occurs all the time. However, if there we find a distinct pattern in their distribution we can suggest it is due to selection. For example, we may find more brown and banded snails in the woodland, and pink and unbanded snails in the open grass, due to differing preferences of the type of niche they live in. It is important to note that selection and drift may not be the only genetic process acting upon their distribution and other genetic and or environmental process may be of an influence.
Definitions: Genetic drift: a random process by which allele frequencies change over time. Selection: a process in which environmental and genetic processes determine which types of organisms thrive better than others, regarded as a factor in evolution. Gene flow: the transfer of genes or alleles from one population to another. Polymorphism: the occurrence of different forms among members of a population or colony. Allele frequency: the proportion of a particular allele in a population.
Hypotheses
Null Hypothesis: The allele frequencies will be the same in the different populations at samples zones 1-6.
Alternative Hypothesis: The allele frequencies will be different in the different populations at samples zones 1-6.
Method
Place a quadrant at each of the 3 sampling sites at both the grassland and the woodland. Count the number of individuals at each sample site (up to 30 individuals). Record the phenotypes of each individual in each of the six populations and create a table with corresponding phenotype categories to record them in. Compare the number of times the phenotypes of populations 1,2,3 occur with the number of times they occur in populations 4,5,6.
Changes to method:
1. Sampling included hedges, grass and woodland instead of just woodland and grassland. This is a better representation of the entire habitat by including all environmental categories.
2. Sampling at the same height: In order to combine two sample populations from the same type of environment for comparison we need to make sure they are not significantly different from one another. They must be distant enough from each other so that it is less likely that their similarities are due to strong gene flow between the two populations or because the individuals are closely related. Since height could affect population size or act as a selective pressure we sampled at the same height. By choosing to place one of the two sample sites of a given environment up to 10 m above a chosen height (represented by the transect line) and the other up to 10m below the chosen height we gave enough flexibility to choose sample sites that are far apart whilst still keeping the height fairly consistent.
Results
Figure 1a:
Table quantifying proportion of colours present in woodland, hedgerow and grassland
Woodland (Population1&6)
Hedgerow (Population2&4)
Grassland (Population3&5)
Totals
Brown
19
1
4
24
Pink
11
3
11
25
Yellow
18
17
31
66
Totals
48
21
46
Figure 1b:
Using the Chi squared contingency calculation we were able to carry a statistical analysis of our results. Using our data of the colour recordings in the different populations, the value of the chi squared is 20.74 which for our degrees of freedom, is 4. Using a 0.05 value and our degrees of freedom, the critical value is 9.488. Our chi-squared value exceeds this and so we reject the null hypothesis, that there is no significant difference in phenotypic variation between the populations at 1 and 6 (Woodland), 2 and 4(Hedgerow) and 3 and 5 (Grassland). Our results gave us a p value of 0.0004 showing that if in fact it was the case that our null hypothesis was actually correct; there would be a 0.4% chance of getting the results in our data or more extreme.
Figure 1c:
Percentage Deviations
Woodland
Hedgerow
Grassland
Pink
+89.7%
-77.2%
-58.3%
Brown
+5.4%
-34.3%
+10%
Yellow
-34.7%
+41.1%
+17.4%
The percentage deviation table shows that the pink shell phenotype in the woodland has the greatest deviation with being 87.9% larger than its expected value and the brown shell phenotype in the woodland had the smallest deviation at 5.4%.
Figure 2a:
Standardized residuals
Woodland
Hedgerow
Grassland
Brown
+2.84
-1.62
-1.81
Pink
+0.17
-0.73
+0.32
Yellow
-1.82
+1.43
+0.9
Figure 2b:
Table quantifying proportion of banding present in woodland, hedgerow and grassland
Woodland (Population1&6)
Hedgerow (Population2&4)
Grassland (Population3&5)
Totals
0
13
12
0
15
1-2
10
2
3
15
3
5
9
15
29
4-5
20
8
27
55
Total
48
21
45
Figure 2c:
This time using our recorded banding data we performed the same statistical analysis for the populations giving a chi squared value of 27.04. Using our degrees of freedom, 6, and a 0.05 value the critical region 12.952, our chi-squared value exceeds this and so we reject the null hypothesis. If the null hypothesis was however correct there would be a 0.1% chance of getting these data.
With respects to both colour and banding the contingency table shows that there is a significant difference in the populations that is very unlikely to be due to sampling variation, the value of chi-squared cause us to reject the null hypothesis on both levels, supporting the alternate hypothesis that there are differences in allele frequencies between the different sample sites: Woodland, Hedgerow and Grassland.
Figure 2d:
Percentage deviations
Woodland (Population1&6)
Hedgerow (Population2&4)
Grassland (Population3&5)
0
+105.8%
-27.6%
-100%
1-2
+58.5%
-27.6%
-49.3%
3
-59.1%
+68.5%
+31%
4-5
-13.6%
-21%
+24.4%
Standardised residuals:
Woodland (Population1&6)
Hedgerow (Population2&4)
Grassland (Population3&5)
0
+2.26
-0.46
-2.43
1-2
+1.47
-0.46
-1.2
3
-2.06
+1.58
+1.05
4-5
-0.66
-0.67
+1.14
Discussion Our results show that there was a significant difference in both the colour of shell & number of bands on the shell between populations in the woodland (population 1&6), hedgerow (populations 2&4) and grassland (populations 3&5). For example, by referring to figure 2b, the woodland contained the highest frequency of zero banded snails (13) compared to the grassland (0). This therefore supports our hypothesis that there will be differences in allele frequencies between the populations at each sampling zone. We can therefore reject the null hypothesis which stated that there would be no difference in allele frequencies between the different sample zones.
The fact that we found significant differences between the shell colours and banding of snails at the three different types of sampling site we can suggest that selection is occurring. Jones et al (1977) indicated how different types of selection can affect the allele frequencies in snail populations. Visual selection refers to selection that occurs due to predation. Snails with certain visual characteristics are often less likely to be predated than those who exhibit a less desirable phenotype. For example, our results showed a high number of yellow snails in the grassland populations. It is likely that the yellow snails can avoid predation by blending into the green-yellow backdrop of this environment much better than their pink or brown counterparts. These findings are supported by Jones et al (1977) where it was stated that in green and variegated backgrounds yellow banded snails predominate. Likewise, the increased number of brown snails in the woodland areas when compared with grassland and hedgerow suggests visual selection in favour of brown snails when the background is dark and woody.
It is also possible to explain our results using climatic selection, which suggests that the differences in banding and colour in shells may alter thermal regulation and absorption of solar energy. Lamotte (1959) suggested that we would expect to find a higher proportion of yellow snails in grassland due to the fact that yellow snails are better at withstanding insolation and extreme temperatures than snails exhibiting the pink and brown phenotypes. This may therefore be another way of explaining our results.
The fact that our results can be explained through various types of selection, which is supported by previous literature, suggests that selection does play a role in the polymorphism of C. nemoralis.
However, it is important to note that genetic drift is a process that is happening all the time and can also affect the abundance and distribution of the various types of C. nemoralis. For example, snails with three bands are roughly found in equal proportions in all three types of sample areas. This exemplifies that drift is continuously occurring despite selection pressures.
To conclude, we believe that both genetic processes are occurring in the sample site and are affecting the distribution and abundance of C. nemoralis. It is difficult to deduce that one process cannot occur without the other having an effect.
To further improve the investigation, we could have undertaken more samples in each population. For example, three samples in each site. This would make the study more reliable and accurate.
References
Jones, J.S., Leith, B.H and Rawlings, P. 1977. Polymorphism in Cepaea: A problem with too many solutions? Annual Review of Ecology and Systematics, Volume 8, page 109-143.4
Lamote, H. 1959. Polymorphism of natural populations of Cepaea nemoralis. Cold Spring Harbor Symp. Qyant. Biol., 24, 65-86.
By: Emma Louise Black, Christopher Tobi Travih, Daniel Guenin, Kamela Bellovoda, Joey Kielty.
Introduction
Cepea is a highly polymorphic species of snail. One of the most abundant types of Cepea found at the site is C. nemoralis. The most well-known genetic polymorphism of this species is that of shell pattern. The shell of C. nemoralis has polymorphism for colour and for presence, number and appearance of up to five dark bands. The major classes of colour are yellow, pink and brown. There is much variation in how the bands are expressed which includes, fully pigmented bands, which are common, and interruption or complete loss of the bands. In many of the populations studied, fusions of adjacent bands and spreading of band pigment over the whole of the shell has also been frequently found. Less common are changes in the colour of the band pigment to orange or light yellow. Genetic variation has also been found in shell shape and size, fecundity, viability, developmental time and some aspects of behaviour in some populations (Jones et al 1977).
According to the Jones et al. study on C. nemoralis there are different kinds of selection acting on this species that are very important for the maintenance of polymorphism. There is visual selection which is based on selection by predators. Another form of selection is climate for which the absorption of the energy from the sun varies between dark/light and banded/un-banded shells. There are also frequency dependent selection in which rare alleles are selected over more spread ones. This happens due to the fact that the individuals containing rare morphs are not easily found. Other types of selection important for the maintenance of shell polymorphism are density dependent and disruptive selection (Jones et al 1977).
Genetic drift and selection may alter the distribution and abundance of the different types of C. nemoralis. Our experimental design may distinguish the effects of drift from selection. We have sampled in each of the different environments (woodland, open grass and hedges) at heights differing of 10 metres above and below the transect line. If there is random distribution of the different types of snail and there is no distinct pattern in their distribution, then we can suggest it is due to genetic drift as every species is subjected to drift which occurs all the time. However, if there we find a distinct pattern in their distribution we can suggest it is due to selection. For example, we may find more brown and banded snails in the woodland, and pink and unbanded snails in the open grass, due to differing preferences of the type of niche they live in. It is important to note that selection and drift may not be the only genetic process acting upon their distribution and other genetic and or environmental process may be of an influence.
Definitions:
Genetic drift: a random process by which allele frequencies change over time.
Selection: a process in which environmental and genetic processes determine which types of organisms thrive better than others, regarded as a factor in evolution.
Gene flow: the transfer of genes or alleles from one population to another.
Polymorphism: the occurrence of different forms among members of a population or colony.
Allele frequency: the proportion of a particular allele in a population.
Hypotheses
Null Hypothesis: The allele frequencies will be the same in the different populations at samples zones 1-6.
Alternative Hypothesis: The allele frequencies will be different in the different populations at samples zones 1-6.
Method
Place a quadrant at each of the 3 sampling sites at both the grassland and the woodland. Count the number of individuals at each sample site (up to 30 individuals). Record the phenotypes of each individual in each of the six populations and create a table with corresponding phenotype categories to record them in. Compare the number of times the phenotypes of populations 1,2,3 occur with the number of times they occur in populations 4,5,6.
Changes to method:
1. Sampling included hedges, grass and woodland instead of just woodland and grassland. This is a better representation of the entire habitat by including all environmental categories.
2. Sampling at the same height: In order to combine two sample populations from the same type of environment for comparison we need to make sure they are not significantly different from one another. They must be distant enough from each other so that it is less likely that their similarities are due to strong gene flow between the two populations or because the individuals are closely related. Since height could affect population size or act as a selective pressure we sampled at the same height. By choosing to place one of the two sample sites of a given environment up to 10 m above a chosen height (represented by the transect line) and the other up to 10m below the chosen height we gave enough flexibility to choose sample sites that are far apart whilst still keeping the height fairly consistent.
Results
Figure 1a:
Table quantifying proportion of colours present in woodland, hedgerow and grassland
Totals
Brown
Pink
Yellow
Totals
Figure 1b:
Using the Chi squared contingency calculation we were able to carry a statistical analysis of our results. Using our data of the colour recordings in the different populations, the value of the chi squared is 20.74 which for our degrees of freedom, is 4. Using a 0.05 value and our degrees of freedom, the critical value is 9.488. Our chi-squared value exceeds this and so we reject the null hypothesis, that there is no significant difference in phenotypic variation between the populations at 1 and 6 (Woodland), 2 and 4(Hedgerow) and 3 and 5 (Grassland). Our results gave us a p value of 0.0004 showing that if in fact it was the case that our null hypothesis was actually correct; there would be a 0.4% chance of getting the results in our data or more extreme.
Figure 1c:
Woodland
Hedgerow
Grassland
Pink
+89.7%
-77.2%
-58.3%
Brown
+5.4%
-34.3%
+10%
Yellow
-34.7%
+41.1%
+17.4%
The percentage deviation table shows that the pink shell phenotype in the woodland has the greatest deviation with being 87.9% larger than its expected value and the brown shell phenotype in the woodland had the smallest deviation at 5.4%.
Figure 2a:
Figure 2b:
Figure 2c:
This time using our recorded banding data we performed the same statistical analysis for the populations giving a chi squared value of 27.04. Using our degrees of freedom, 6, and a 0.05 value the critical region 12.952, our chi-squared value exceeds this and so we reject the null hypothesis. If the null hypothesis was however correct there would be a 0.1% chance of getting these data.
With respects to both colour and banding the contingency table shows that there is a significant difference in the populations that is very unlikely to be due to sampling variation, the value of chi-squared cause us to reject the null hypothesis on both levels, supporting the alternate hypothesis that there are differences in allele frequencies between the different sample sites: Woodland, Hedgerow and Grassland.
Figure 2d:
Standardised residuals:
Discussion
Our results show that there was a significant difference in both the colour of shell & number of bands on the shell between populations in the woodland (population 1&6), hedgerow (populations 2&4) and grassland (populations 3&5). For example, by referring to figure 2b, the woodland contained the highest frequency of zero banded snails (13) compared to the grassland (0). This therefore supports our hypothesis that there will be differences in allele frequencies between the populations at each sampling zone. We can therefore reject the null hypothesis which stated that there would be no difference in allele frequencies between the different sample zones.
The fact that we found significant differences between the shell colours and banding of snails at the three different types of sampling site we can suggest that selection is occurring. Jones et al (1977) indicated how different types of selection can affect the allele frequencies in snail populations. Visual selection refers to selection that occurs due to predation. Snails with certain visual characteristics are often less likely to be predated than those who exhibit a less desirable phenotype. For example, our results showed a high number of yellow snails in the grassland populations. It is likely that the yellow snails can avoid predation by blending into the green-yellow backdrop of this environment much better than their pink or brown counterparts. These findings are supported by Jones et al (1977) where it was stated that in green and variegated backgrounds yellow banded snails predominate. Likewise, the increased number of brown snails in the woodland areas when compared with grassland and hedgerow suggests visual selection in favour of brown snails when the background is dark and woody.
It is also possible to explain our results using climatic selection, which suggests that the differences in banding and colour in shells may alter thermal regulation and absorption of solar energy. Lamotte (1959) suggested that we would expect to find a higher proportion of yellow snails in grassland due to the fact that yellow snails are better at withstanding insolation and extreme temperatures than snails exhibiting the pink and brown phenotypes. This may therefore be another way of explaining our results.
The fact that our results can be explained through various types of selection, which is supported by previous literature, suggests that selection does play a role in the polymorphism of C. nemoralis.
However, it is important to note that genetic drift is a process that is happening all the time and can also affect the abundance and distribution of the various types of C. nemoralis. For example, snails with three bands are roughly found in equal proportions in all three types of sample areas. This exemplifies that drift is continuously occurring despite selection pressures.
To conclude, we believe that both genetic processes are occurring in the sample site and are affecting the distribution and abundance of C. nemoralis. It is difficult to deduce that one process cannot occur without the other having an effect.
To further improve the investigation, we could have undertaken more samples in each population. For example, three samples in each site. This would make the study more reliable and accurate.
References
Jones, J.S., Leith, B.H and Rawlings, P. 1977. Polymorphism in Cepaea: A problem with too many solutions? Annual Review of Ecology and Systematics, Volume 8, page 109-143.4
Lamote, H. 1959. Polymorphism of natural populations of Cepaea nemoralis. Cold Spring Harbor Symp. Qyant. Biol., 24, 65-86.